British Gynaecological Cancer Society (BGCS) Epithelial Ovarian .

The American Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Randomised Controlled Trial demonstrated that screening asymptomatic postmenopausal women with a single threshold value of CA125 does not result in reduction of mortality, despite 13 years of long term follow up. Diagnostic evaluation following a false-positive screening test result was associated with complications.(5, 6). The UK Collaborative Trial of Ovarian Cancer Screening (UKCTOCS) trial randomised 202,000 women to observation alone, multimodal screening (MMS), with an algorithm based on serial values of CA125 and follow on transvaginal ultrasound scanning (TVS) for abnormal results, or serial TVS alone. The results showed no reduction in mortality in the primary analysis, but a possible reduction in mortality after exclusion of prevalent cases after 7 years of follow-up. Long-term data and costeffectiveness data are awaited.(7) Approximately 1.3% of women in the general population will develop ovarian cancer in their lifetime (4). By contrast, according to the most recent estimates 39% of women who inherit a harmful BRCA1 mutation (5, 6) and 11-17% of women who inherit a harmful BRCA2 mutation will develop ovarian cancer by age 70. (8, 9) The UK Familial Ovarian Cancer Screening Study (UKFOCCS) study evaluated a strategy of annual ultrasound and CA125 measurement in 3,653 women considered at 10% risk of development of ovarian cancer and who declined risk-reducing salpingo-oophorectomy (RRSO). The positive and negative predictive values of incident screening were 25.5% (95% CI, 14.3 to 40.0) and 99.9% (95% CI, 99.8 to 100), respectively. This study is still on-going and work up to 2018 will evaluate a 4-monthly screening strategy with CA125 and ultrasound in this group.(10) RCOG guidelines (2015) did not recommend routine screening in these women idelines/scientific-impact-papers/sip48.pdf). Risk-reducing salpingo-oophorectomy (RRSO) prevents development of epithelial ovarian cancer and reduces mortality in women at high risk for epithelial ovarian cancer (Grade B). Prospective multicentre cohort studies have demonstrated that risk-reducing salpingooophorectomy (RRSO) is associated with a lower risk of EOC, first diagnosis of breast cancer, allcause mortality, breast cancer–specific mortality, and ovarian cancer–specific mortality in BRCA1and BRCA2-mutation carriers, although there still is a residual risk for peritoneal cancer.(11, 12) Ongoing studies are evaluating the role of opportunistic salpingectomy in the prevention of ovarian cancer in low risk women.(13) Tumour markers and Malignancy Indices Tumour markers are not diagnostic tests, but may be helpful in establishing diagnosis and providing baseline values that may be of use during follow up.(14) Prospectively acquired evidence from the United Kingdom Collaborative Trial of Ovarian Cancer Screening Cancer (UKCTOCS) - with 46,237 women triaged using MMS in whom serial CA-125 measurements were interpreted via the risk of ovarian cancer algorithm (ROCA ) - has shown that screening by using ROCA doubles the number of screen-detected EOC compared with a fixed cut off of 35 IU/ml. 7

Caution must be exercised in reassuring women with a single normal CA125 measurement and a focus more on interpreting trends, along with the clinical picture and imaging findings, is likely to define the standard of care in the future.(15) 3. Secondary care and initial pre- treatment assessment In women below 40 years of age with suspected ovarian cancer, measure alpha fetoprotein (AFP), and hCG (human Chorionic Gonadotropin), in addition to CA125, to identify women with non epithelial ovarian lesions (Grade C) . Inhibin should be measured at a presumed diagnosis of a granulosa cell tumor, even though logistically it takes potentially longer to access the results. Secondary care Following referral of a patient with a mass suspicious of ovarian cancer to secondary care, an expansion of the tumour marker panel may facilitate diagnosis. Where CA125 is elevated, a preoperative CA125/CEA ratio 25 , especially in combination with an elevated CA19-9, may indicate peritoneal carcinomatosis from a gastrointestinal tumour and bidirectional gastrointestinal endoscopy should be considered prior to upfront primary debulking surgery.[Grade B] HE4 (human epididymis protein 4) has shown promising diagnostic and prognostic value in triaging younger women, with HE4 not raised in cases of pelvic inflammatory disease and endometriosis despite CA125 elevation being observed. (16-18) Large prospective studies from the International Ovarian Tumour Analysis consortium (IOTA) suggest that using simple “M”(malignant) and “B” (benign) ultrasonographic rules to characterise ovarian masses is highly accurate. Using these simple rules, the reported sensitivity for malignancy was 95%, specificity 91%, positive likelihood ratio 10.37, and negative likelihood ratio 0.06. (19) The accuracy of the IOTA ultrasonographic rules has been demonstrated in secondary care, predominantly with specialists in ultrasonography and their wider use remains under evaluation in the UK (http://www.birmingham.ac.uk/rockets). Results from an on-going study to evaluate the best serum diagnostic tests and ultrasound models to detect ovarian cancer are awaited. Advised examinations prior to deciding treatment In patients with presumed ovarian cancer, radiological staging will provide further information about the extent of disease and potential distant metastases or secondary cancers. (Grade C) CT prediction of suboptimal cytoreduction is not sufficiently reliable and in the absence of favourable data from larger, prospective trials should not be used alone to decide management. (Grade B) MRI should not be routinely used for assessing women with suspected ovarian cancer outside of clinical trials, but can be useful where the results of the USS are not helpful in confirming a 8

diagnosis, especially in young women with a solitary pelvic mass who want a fertility sparing approach. (Grade B) PET CT is not recommended for routine preoperative staging in the NHS outside a clinical trial. (Grade C) CT imaging of the thorax, abdomen and pelvis is recommended to help define the extent of disease and to aid in surgical planning. However, retrospective data have shown that CT cannot accurately predict fine nodule peritoneal carcinomatosis, and therefore mitigate against suboptimal cytoreduction, and that it is not always reliable and reproducible.(2, 20) Current prospective imaging trials are underway to prospectively assess the predictive value of novel imaging techniques in determining operability. CT has significant value in excluding distant macroscopic disease spread, including intraparenchymal liver or lung metastases and retroperitoneal node involvement, and in excluding synchronous cancers from other sites or thromboembolic events that may alter management. (Grade B) Current national guidance recommends that MRI should not routinely be used for assessing women with suspected ovarian cancer, but may be used as a problem-solving tool and adjunct to other imaging modalities. There is also no evidence based value in the routine use of specialized imaging techniques such as positron emission tomography–computed tomography (PET CT), although it may be useful as a problem-solving tool in highly specialised situations (for example in the evaluation of thoracic/mediastinal lymph nodes where secondary intra-abdominal debulking for relapsed disease is under consideration).(21) Diffusion weighted MRI may have a future role in the description of tumour dissemination patterns and assessment of operability, but prospective evidence data for that are warranted.(22) Cytological/Histological Diagnosis If offering cytotoxic chemotherapy to women with suspected advanced ovarian cancer first obtain a confirmed tissue diagnosis by histology in all but exceptional cases. (Grade C)(2) Only commence cytotoxic chemotherapy for suspected advanced ovarian cancer on the basis of positive cytology alone and imaging and without histological confirmation in exceptional cases and where obtaining a tissue sample would be inappropriate. A discussion of such cases at the multidisciplinary team meeting including a careful consideration of the risks and benefits should be documented (Grade C). All patients with histology / cytology showing suspected or actual carcinoma of gynaecological origin should be reviewed at a gynaecology multidisciplinary team (MDT) meeting. 9

Histological diagnosis is not mandatory prior to upfront debulking surgery if the clinical picture, imaging and tumour marker profile are highly suggestive of epithelial ovarian cancer (CA125:CEA ratio 25:1). If ascites is sent for cytological analysis, the absence of malignant cells does not exclude ovarian malignancy, especially in the presence of inflammation (Grade B).(23, 24) The use of immunohistochemistry on a cell block can be of help in such cases if sufficient atypical cells are present to allow for separation from background cells and interpretation of patterns of staining. This is of high value to aid tissue diagnosis in mixed or undifferentiated tumours. Where upfront cytotoxic chemotherapy is offered to women with suspected advanced ovarian cancer, histological tissue diagnosis via image guided biopsy or laparoscopy is mandatory in all but exceptional cases. Cytology alone, together with a CA125/CEA ratio of 25:1 may be sufficient in patients with poor performance status (PS 3,4) and where biopsy is not feasible. (Grade C) In the majority of the cases tissue can be safely obtained through image guided biopsy. The value of laparoscopy in the assessment of operability and impact on overall surgical and clinical outcome of advanced ovarian cancer has not been established in prospective randomised trials . Emerging research protocols utilize laparoscopically obtained multiple intra-abdominal biopsies to define molecular biological profile of each individual patient but the survival benefit of this approach has not been proven in any prospective randomised trials. The routine use of laparoscopy to obtain pre-treatment histology and to assess the operability of disease is not recommended. (Grade B) Data to support laparoscopic assessment to determine tumour resectability is limited and suffers from verification bias.(2) In a Cochrane review, assessing the accuracy of laparoscopy to determine tumour resectability in ovarian cancer, only two studies performed laparoscopy and laparotomy in all patients. (25) The other studies only performed a laparotomy when it was thought that an optimal result was feasible. It is therefore not possible to draw definitive conclusions about the sensitivity of laparoscopy. Three studies developed or validated a prediction model including laparoscopy. Using a prediction model did not increase the sensitivity and resulted in more patients undergoing suboptimal surgery. A multidisciplinary discussion within a quorate MDT as constituted along national guidelines is fundamental to the appropriate management of each individual patient and should be documented prior to a decision to operate, offering chemotherapy or palliative treatment in all but exceptional cases, such as emergency presentations between meetings, and the management of these cases should be agreed and described in a departmental gynaecological cancer operational document. Significance and caveats of cytology In about two thirds of patients with known ovarian carcinoma, malignant cells are seen in the ascitic fluid. However, there are strong reservations about using peritoneal or ascitic cytology without 10

histological confirmation in the primary diagnosis of ovarian cancer. Cytological preparations lack architectural patterns and false positive tests may be obtained from serous borderline tumours and from exfoliation of other cells, such as epithelial cells from Müllerian rests and reactive mesothelial cells, which may be mistaken for carcinoma. This problem may be partially resolved through constructing cell blocks and performing appropriate immunohistochemistry, but despite this, the use of cytology in the diagnosis of ovarian carcinoma has a high false negative rate and is operator dependent. Histological confirmation is recommended prior to treatment with chemotherapy. In exceptional cases, where obtaining material for histology is not possible or is associated with a high risk due to the poor performance status or co-morbidities of the patient, cytology may be used alone in establishing a pre-chemotherapy diagnosis. In women with pleural effusions, aspiration and examination for malignant cells and cytology should be considered to confirm staging (preferably with immunohistochemistry on cell block). (23, 24). When used in trial settings, cytological preparations are suboptimal for archiving, tissue microarrays and some molecular testing. 4. Pathology and genetics The provision of a minimum set of clinical information on the histopathology request form is crucial to ensure a histopathology report of high enough quality for the accurate diagnosis and appropriate management. (Grade D) Frozen section may be performed, if the result will alter the intra-operative management although there are limitations to the technique. (Grade B) Clinical information required on the specimen request form The Royal College of Pathology guidelines for reporting ovarian carcinomas mandate the provision of minimum clinical details to include demographics, clinical presentation, results of previous biopsies, radiological investigations for tumour staging, and details of the surgical procedures performed. It is desirable to include details of any family history of cancer and relevant hormonal therapy. The nature of surgical specimens from multiple sites should be carefully recorded and the specimen pots labelled to correspond to the specimen details on the request form and appropriately labelled as to site of origin. Primary site assignment The origin of high-grade serous ovarian carcinoma (HGSC) has been the subject of intense study. The distal fallopian tube has emerged as the likely site of origin for most HGSC. (26) This observation is, in great part, attributable to the use of sampling protocols that thoroughly examine the distal fallopian tube and also due to the greater number of specialist pathologists with a sub-specialty interest in gynaecological pathology. The discovery of serous tubal intraepithelial carcinoma (STIC) in women with BRCA1 or BRCA2 mutations following risk-reducing salpingo-oophorectomies (RRSO) and in women with advanced ovarian carcinoma lead to the hypothesis that the natural history of pelvic 11

HGSC might involve an origin in most cases of the distal fimbria of the fallopian tube. Identification of STIC in 18% to 60% of cases of advanced/symptomatic HGSC supports this assertion. STIC lesions are characterized by DNA damage, TP53 mutation, and progressive molecular abnormalities that are also seen in high-grade serous carcinoma. An origin from epithelial inclusion cysts in the ovary has been proposed as a potential explanation as site of origin in the cases where complete examination of the fallopian tube does not reveal STIC. A consensus statement on primary site assignment in tuboovarian HGSC has been made. (27) Immunohistochemical features of HGSC HGSC of tubo-ovarian and peritoneal origin have similar morphological and immunohistochemical features. HGSC can be arra

development of ovarian cancer (Grade A) The role of ovarian cancer screening in women at high risk of ovarian cancer has yet to be established (Grade B) Clinical examination and serum CA125 measurement should be considered in women with symptoms suggestive of ovarian cancer. If the A125 is 35 IU/ml, or if a pelvic mass or other